Within the feedback loop, the sensor provides a measured output that can be compared with the reference value. The resulting deviation gives the controller a basis for changing the system input, so the response opposes the disturbance rather than allowing the output to remain displaced. This interaction supports accurate operation.
Feedforward control uses a measured disturbance to adjust the input before that disturbance affects the output. Its timing distinguishes it from feedback, which responds after a deviation is detected. In a disturbance-rejection design, feedforward can act earlier, while feedback addresses remaining effects and deviations in the controlled system.
Model uncertainty matters because the system may not behave exactly as expected, especially when operating conditions change. Disturbance rejection must therefore be considered in terms of robustness, meaning dependable performance despite those variations. A suitable design aims to preserve the desired output while maintaining stability and accuracy across encountered conditions.
An engineering workflow begins by specifying the desired output and reference, then using sensors to obtain the measured output. The controller compares these values and adjusts the input in response to the deviation. If the disturbance can be measured before it affects the system, engineers may add feedforward compensation to act in advance.
Disturbance rejection applies across several engineering systems, including motor drives, aircraft control, and industrial temperature regulation. In each case, unwanted influences or operating changes can affect the desired output, making controlled compensation important. These examples show that the principle supports both dynamic systems, such as aircraft, and industrial process regulation.
Engineers can evaluate the approach by considering whether the system maintains stability, accuracy, and robustness when disturbances or operating conditions change. Reliable performance means the desired output remains controlled despite external influences, internal variations, or model uncertainty. These criteria connect disturbance rejection directly to dependable operation in practical engineering systems.